Cement and Concrete Research 31 (2001) 1867 – 1872

Effect of curing temperature and type of cement on early-age shrinkage of

high-performance concrete
Pietro Luraa,*, Klaas van Breugela, Ippei Maruyamab
a
Concrete Structures Group, Stevin Laboratory, Faculty of Civil Engineering and Geosciences, Delft University of Technology, P.O. Box 5048, 2600 GA Delft,
The Netherlands
b
Building Materials Laboratory, Faculty of Architecture, Tokyo University, Tokyo, Japan
Received 17 January 2001; accepted 29 June 2001

Abstract

This paper presents the results of an experimental study on the influence of curing temperature and type of cement [Portland cement and
blast-furnace slag (BFS) cement] on the autogenous deformations and self-induced stresses in early-age concrete. It was found that higher
temperatures do not lead to higher deformations in the observed period, but generally cause a faster shrinkage and a faster development of
self-induced stresses. Another experimental finding is that, at the temperatures tested, concrete made with BFS cement shows higher
shrinkage in the first days than concrete made with Portland cement. D 2001 Elsevier Science Ltd. All rights reserved.

Keywords: Temperature; Shrinkage; Granulated blast-furnace slag; Portland cement

1. Introduction On the other hand, at the scale of the concrete structure,

In high-performance concrete (HPC) mixtures, a low
water/binder (w/b) ratio and addition of silica fume cause
a significant drop of the relative humidity during hydration.
As a consequence of self-desiccation, the cement paste
undergoes shrinkage. Since the Young’s modulus of the
aggregates is higher than that of the hardening paste,
autogenous shrinkage will result in tensile stresses within
the cement paste and in bulk deformation of the concrete
itself. Both these phenomena should be avoided as much as
possible since they could induce micro- or macrocracking
and impair the concrete quality.
In particular, internal restraint, due to the presence of the
aggregates in the mixture, may cause microcracking. Many
authors have predicted the occurrence of these cracks with
analytical [1] or numerical methods [2]. The actual detec-
tion of the shrinkage microcracks using microscope obser-
vations is more difficult, due to problems with the sample
preparation [3].
* Corresponding author. Tel.: +31-15-278-4580; fax: +31-15-278-
5895.
E-mail addresses: p.lura@citg.tudelft.nl, p.lura@ct.tudelft.nl (P. Lura).
autogenous shrinkage, added to temperature-induced defor-
mations, may lead to surface cracks and also to through-
cracks. Even in the case of isothermal sealed curing, where
autogenous shrinkage is the only cause of deformation, self-
induced tensile stresses caused by external restraint may
induce cracking of the concrete [4 – 6], potentially compro-
mising the durability.
Measurements of autogenous deformation are normally
performed at room temperature. Up to now, only a few
authors have investigated the effect of different curing
temperatures on autogenous shrinkage. Most of these tests
focus on cement pastes [7– 9]. From these studies, a rather
unsystematic temperature dependency of the autogenous
deformation emerges. According to Ref. [6], autogenous
deformation at different curing temperatures cannot be
predicted only with a maturity function. Results in Ref.
[10], on the other hand, show a more systematic behavior,
and the authors conclude that autogenous deformation can
be simulated with the maturity concept, provided that also a
further temperature-correcting factor is applied.
Most of the experimental research on autogenous shrink-
age has been concentrated on Portland cement mixtures. In
some European countries, blast-furnace slag (BFS) cement
has been used for more than a century. In particular, in the
Netherlands, BFS cements (with slag contents of up to 70%)

0008-8846/01/$ – see front matter D 2001 Elsevier Science Ltd. All rights reserved.
PII: S 0 0 0 8 - 8 8 4 6 ( 0 1 ) 0 0 6 0 1 - 9
1868 P. Lura et al. / Cement and Concrete Research 31 (2001) 1867–1872

have been widely and successfully applied, particularly in Table 2

Chemical composition of the cements used
marine structures [11]. Advantages of this kind of cement
include environmental friendliness (due to the reuse of a CEM I 52.5 R CEM III/B 42.5 LH HS
(Portland) (BFS cement)
waste material), low hydration heat and a finer pore struc-
Blaine  530 m2/kg Blaine  490 m2/kg
ture, which improves water tightness [12] and durability.
CaO 64% 46%
Since both self-desiccation and autogenous deformation
SiO2 21% 30%
depend on the chemical composition of the cement [13] Al2O3 5% 10%
and on the pore size distribution of the cement paste [14], it Fe2O3 3% 1.5%
is expected that BFS cement mixtures will show different MgO 2% 8%
shrinkage behavior than Portland cement mixtures. In fact, SO32 + 3.3% 3%
Cl  0.05%  0.03%
higher shrinkage values for BFS mixtures have been
Slag content –  70%
reported by different authors [15 – 17]. Loss on ignition  1% –
Insoluble residue  1%  1%
Na2O-equiv. 0.6% 0.5%

2. Materials and methods

the plates and the insulating material. Temperature differ-
In the present research, the effects of different curing ences inside the concrete could be kept as low as 1.5 K.
temperatures and cement types on autogenous deformation Length changes of the hardening concrete were measured
of HPC mixtures were investigated. Four different curing with two external quartz rods provided with linear variable
temperatures (10, 20, 30 and 40 °C) were imposed. Also the differential transducers (LVDTs) at both ends. The quartz
type of cement (Portland cement, BFS cement and a blend rods were connected to steel bars cast in the concrete and
of the two) was varied, in order to assess its influence on the measured displacements over a length of 750 mm. The rods
early-age deformations. For every concrete mixture and could be fixed to the cast-in bars when the concrete had
curing temperature, the self-induced stresses in the case of reached sufficient strength to support them (see Fig. 1a).
total restraint were measured and the occurrence of cracking After casting, the top surface of the concrete was covered
was registered. with a tight cover in order to avoid moisture loss to the
Mixtures with w/b ratio of 0.35 and 5% silica fume were environment. The autogenous deformation up to 6 days after
considered (see Table 1). Portland cement (CEM I 52.5 R) casting was recorded.
was used in Mixture A, BFS cement (CEM III/B 42.5 LH
HS) in Mixture B and a blend of the two in Mixture C.
Details about the chemical composition of the cements are
reported in Table 2. It is noticed that the slag content of the
BFS cement amounts to about 70%, and the Blaine fineness
of the two cements is 490 m2/kg for the BFS cement and
530 m2/kg for the Portland cement. Data about the perfor-
mance of Mixture C cured at 20°C were obtained in a
previous research study [5]. For this particular mix and
curing conditions, the Young’s modulus was not measured.
Measurements of the free deformations were performed
with an Autogenous Deformation Testing Machine (ADTM)
[5,14]. The concrete was cast in a prismatic mould,
1000  150  150 mm3, made with thin steel plates pro-
vided with an external insulating material. The mould could
be cooled or heated by a system of tubes located between

Table 1
Mixture compositions of concrete with w/b ratio of 0.35
Mixture composition (kg/m3) A B C
CEM I 52.5 R (Portland cement) 475.0 – 238.0
CEM III/B 42.5 LH HS (BFS cement) – 475.0 237.0
Water (including water in admixtures) 175.8 175.8 175.8
Crushed aggregate (4 – 16 mm) 944.2 944.2 944.2
Sand 0 – 4 mm 772.5 772.5 772.5 Fig. 1. (a) Detail of experimental setup for the measurement of the free
Lignosulphonate 0.9 0.9 0.9 deformations of hardening concrete (ADTM) and (b) top view of
Naphtalene sulphonate 8.1 7.1 7.6 experimental setup for the determination of stress development in hardening
Silica fume slurry (50% powder, 50% water) 50.0 50.0 50.0 concrete (TSTM).
 P. Lura et al. / Cement and Concrete Research 31 (2001) 1867–1872 1869

The self-induced stresses were measured with a Thermal Table 4

Young’s modulus of concrete with w/b ratio of 0.35, sealed curing at
Stress Testing Machine (TSTM) [5]. This device consisted
different temperatures
of a steel frame (for details and dimensions, see Fig. 1b)
Young’s modulus (GPa)
supporting a mould in which the concrete was cast. The
specimens had a prismatic shape with dovetailed heads at Temperature 1 2 3 7 14 28 56
Mixture (°C) day days days days days days days
the ends. Two rigid steel claws held the specimen and were
able to exert a tensile or a compressive force. The imposed A 10 – 30.1 32.2 – 34.5 35.5 –
20 30.6 33.7 – – 35.7 38.2 –
strain on the concrete in the TSTM was deduced from the
30 31.4 33.0 33.9 – 37.4 – –
free deformations measured in the ADTM; the control of the 40 32.0 35.2 35.7 – 39.7 – –
device was fully automatic. In this research, only stresses B 10 – 6.7a 18.4b – 29.3 33.5 –
obtained under total restraint were measured. 20 11.7 24.2 27.3 – – 36.7 –
For the same mixtures, the compressive strength and the 30 27.4 32.3 32.6 – 35.5 – –
40 28.3 33.5 33.7 – – – 38.7
Young’s modulus at different ages were also tested, on
C 10 – 24.9a 26.5b 32.7 – 37.3 –
sealed specimens cured at the four different temperatures. 30 27.8 32.2 32.2 – 39.5 – –
The compressive strength was measured on concrete cubes, 40 30.7 34.5 34.2 – 38.0 – –
150  150  150 mm3; the Young’s modulus in compres- a
Tested 1.5 days after casting.
sion was tested on prisms, 100  100  400 mm3. Cubes b
Tested 2.5 days after casting.
and prisms were cast in temperature-controlled steel moulds.
The specimens tested at later ages were removed from the
moulds after 6 days, sealed with plastic and aluminum foils In Figs. 2 – 4, the autogenous deformations and self-
and stored at constant temperature until the moment induced stresses of Mixtures A, B and C are plotted. The
of testing. autogenous deformations were zeroed at the time when
stresses were first recorded in the TSTM. Thus, only the
stress-inducing deformations are shown. For Mixtures A
and C, this means only shrinkage, since only tensile stresses
3. Results

Results of compressive strength and E-modulus are

reported in Tables 3 and 4. The results represent the average
value of three specimens. Mixture A, made with Portland
cement, shows, as expected, both the fastest strength gain
and the highest value at 28 days, followed by Mixture C.
The strength of Mixture B develops slowly, especially at
the lower temperatures. Similar trends are found for the
Young’s modulus.

Table 3
Cube compressive strength of concrete with w/b ratio of 0.35, sealed curing
at different isothermal temperatures
Mean cube compressive strength (MPa)
Temperature 1 2 3 7 14 28 56
Mixture (°C) day days days days days days days
A 10 – 57.2 65.8 81.7 91.7 97.7 –
20 53.2 65.1 – 85.9 94.0 99.5 –
30 55.4 68.5 78.6 92.3 100.1 – –
40 66.5 74.2 86.6 92.5a 97.0 – –
B 10 – 5.2b 13.1c 34.5 51.8 58.9 –
20 4.9 20.4 32.7 53.8 – 65.9 –
30 23.6 44.6 50.6 59.6 64.8 – –
40 31.8 47.7 53.6 59.4a – – 69.7
C 10 – 31.1b 38.1c 57.4 72.2 82.8 –
20 42 52 58 70 – 82 –
30 40.7 54.5 63.9 – 84.8 – –
40 44.9 65.1 71.4 78.8a 79.2 – –
a
Tested 8 days after casting. Fig. 2. (a) Autogenous deformation and (b) self-induced stresses of Mixture
b
Tested 1.5 days after casting. A (see Table 1). Isothermal sealed curing at different temperatures (10, 20,
c
Tested 2.5 days after casting. 30 and 40°C). Shrinkage is plotted as negative.
1870 P. Lura et al. / Cement and Concrete Research 31 (2001) 1867–1872

problems. The other specimens showed maximum stresses

between 1.5 and 2 MPa after 6 days. No specimen cracked
in the testing period. In order to compare the data obtained
at different temperatures, the degree of hydration of the
mixes was calculated with the program HYMOSTRUC
[18]. The degree of hydration at 5 days from casting was
0.58 for 10°C curing, 0.63 for 20°C curing, 0.65 for 30°C
curing and 0.69 for 40°C curing.
In Fig. 3, the results of Mixture B, made with BFS
cement, are shown. Also in this case, the influence of the
curing temperature on the autogenous deformation is uncer-
tain. The initial swelling of the specimens cured at 20, 30
and 40°C was stress-inducing, with compressive stresses
lower than 0.3 MPa. The following shrinkage occurred
earlier for the specimen cured at 40°C, but 30 h after casting
the shrinkage of the specimen cured at 30°C became higher.
Shrinkage of the specimen cured at 10°C developed slowly,
but at the end of the test it was about the same as in the case
of 40°C curing. Shrinkage values after 6 days were about
170 – 180 microstrain for curing at 10, 20 or 40°C, and about
230 microstrain for curing at 30°C. These values are
noticeably higher than those found for Portland cement
mixtures (see Fig. 2). Also in this case, the self-induced

Fig. 3. (a) Autogenous deformation and (b) self-induced stresses of Mixture

B (see Table 1). Isothermal sealed curing at different temperatures (10, 20,
30 and 40°C). Shrinkage is plotted as negative. A cross indicates failure of
the specimen.

were measured in the stress-rig. For Mixture B, at some

curing temperatures, shrinkage was preceded by swelling,
inducing a low ( < 0.3 MPa) compressive stress in the
TSTM. The stress-inducing deformations were preceded
by a phase, lasting for some hours, where the concrete
deformed plastically without generating any stress in the
TSTM. In this plastic phase, all the concrete mixtures
expanded shortly after the beginning of the measurements,
with swelling up to 50 microstrain. For example, Fig. 5
shows the complete record of the deformations in the first
2 days in the case of 40°C curing.
The results of Mixture A, regarding stress-inducing
deformations and stress development, are shown in Fig. 2.
The stress-inducing shrinkage developed rather unsystema-
tically with temperature. For example, the specimen cured
at 40°C showed a faster shrinkage in the first hours, but
then it slowed down. After 1 day, the total value was less
than for 20°C curing. Shrinkage values after 6 days varied
between 130 and 170 microstrain for the four curing
temperatures. The self-induced stresses in case of totally
Fig. 4. (a) Autogenous deformation and (b) self-induced stresses of Mixture
restrained deformations increased with temperature C (see Table 1). Isothermal sealed curing at different temperatures (10, 20,
increase. The recording of the stresses at 40°C curing was 30 and 40°C). Shrinkage is plotted as negative. A cross indicates failure of
stopped at about 30 h after casting, due to technical the specimen.
 P. Lura et al. / Cement and Concrete Research 31 (2001) 1867–1872
Fig. 5. Autogenous deformations of Mixtures A, B and C in the first 48 h
after casting, cured at 40°C. All the measured deformations are shown.
stresses gave a clearer picture, with higher stresses at higher
curing temperatures. It should be noticed that the specimen
cured at 30°C cracked 2 days after casting, when the
measured stress was about 2.5 MPa. The specimen cured
at 40°C probably experienced microcracking around that
age, a fact that is shown by the decreasing slope of the stress
curve, while the deformations were still increasing. In this
case, the calculated degree of hydration after 5 days ranged
from 0.4 for 10°C curing to 0.58 for 40°C curing.
Results of tests on Mixture C, made with blended
cement, are shown in Fig. 4. Shrinkage was lowest for
10°C and highest for 40°C curing. The specimen cured at
20°C showed faster shrinkage development than the one
cured at 30°C. It should be pointed out, however, that the
data at 20°C curing had been obtained in a previous research
[5]; the materials used might have been slightly different
(different cement batches, for example). Shrinkage values
after 6 days were 120 microstrain for 10°C curing, 140
microstrain for 30°C and 210 microstrain for 20 or 40°C.
These values lie between the ones found for Mixtures A and
B. Self-induced stresses were higher for curing at 40 and
30°C. The specimen cured at 30°C cracked 3 days after
casting, with tensile stresses of about 2.5 MPa. The speci-
men cured at 40°C did not crack in the testing period,
reaching stresses higher than 3 MPa. The stresses of the
specimen cured at 10°C were extremely low, below 1 MPa
after 6 days. For this concrete, the calculated degree of
hydration after 5 days ranged from 0.53 for 10°C curing to
0.68 for 40°C curing.
4. Discussion and conclusions
From the analysis of the experimental results on three
concrete mixtures and four different curing temperatures,
the following conclusions can be drawn.
Comparison of the results for the same mixture at
different curing conditions shows that, although higher
temperatures improve the initial strength development, the
1871
value at a later stage, 14 or 28 days, seems to be penalized,
as already observed by other authors [19]. For example,
comparing specimens of Mix A and C cured at 30 and
40°C, it can be noticed that specimens cured at the higher
temperatures show a lower strength at 14 days.
The Young’s modulus development shows trends similar
to the compressive strength for the three mixes but, as
already observed [20], the stiffness increases faster than
the strength.
The phenomenon of the very early-age swelling is well
known in literature, and seems to occur with concretes of
different w/b ratios and to be independent from the measur-
ing device [4,6,8,15,21], even if the absolute values may
differ quite substantially. A possible reason of this swelling
could be reabsorption of bleeding water. It has been
observed [6] that removing the bleeding water reduces the
swelling, but does not eliminate it totally. The residual
swelling could be due to internal bleeding in the mixture.
In the present research, no external bleeding has been
observed, but the occurrence of internal bleeding cannot
be excluded. Another possible explanation of this phenom-
enon could be found at the scale of the hydration products.
Even if the reaction products have a lower volume (about
7% less) than the reagents, due to their shape they form a
spatial network. Growth of further reaction products inside
the network generates an internal pressure that may cause
moderate swelling of the system [22]. Other authors attri-
bute the swelling to ettringite formation [23].
The effect of a temperature increase on the development
of the autogenous deformations is unsystematic and varies
for different sorts of cement. It seems that higher tempera-
tures do not necessarily lead to higher deformations. This
fact confirms previous findings by other authors [6,8].
Differences in the shrinkage values after 6 days could also
be due to experimental scatter in the results (a maximum of
two tests was run for each temperature, with a difference of
about 8%).
Nevertheless, it can be stated that higher temperatures
generally cause faster development of shrinkage and self-
induced stresses, which might increase the cracking risk. In
fact, none of the specimens cured at 10 or 20°C cracked in
the testing period, while some of the specimens cured at
higher temperature did. While it is not excluded that the
specimens cured at lower temperatures would crack at later
ages, it could be supposed that higher curing temperatures
increase the cracking risk, since the deformations develop at
a higher rate. It has been observed [24] that the rate of the
autogenous deformation is at least as important as its
absolute value, in determining the cracking risk.
Another experimental finding is that, at all tempera-
tures, BFS cement concretes show higher shrinkage after 6
days than Portland concretes. This fact has already been
reported [15 – 17]. The phenomenon could be related to the
supposed denser structure of the BFS cement paste, which
shows smaller pores [15]. These smaller pores could
induce higher capillary forces during the self-desiccation
1872 P. Lura et al. / Cement and Concrete Research 31 (2001) 1867–1872
process, increasing the autogenous shrinkage. The conse-
quences of this fact on the self-induced stresses and the
cracking risk of BFS mixtures still have to be evaluated.
The higher autogenous deformation, in fact, seems not to
result in higher stresses than for Portland cement mixtures,
since the BFS concretes show a later setting time and a
slower stiffness development.
Acknowledgments
The Brite-Euram project IPACS financially supported
this research. The assistance of Mr. E. Horeweg, Mr. A. van
Rhijn, Mr. F.P.J. Schilperoort and Mr. R. Mulder in
performing the experiments is gratefully acknowledged.
References
[1] P. Golterman, Mechanical predictions on concrete deterioration: Part
1. Eigenstresses in concrete, ACI Mater. J. 91 (6) (1994) 543 – 550.
[2] H. Sadouki, F.H. Wittmann, Shrinkage and internal damage induced by
drying and endogenous drying, Proc. Shrinkage 2000 — Int. RILEM
Workshop on Shrinkage of Concrete, Paris, RILEM Publications
S.A.R.L. Cachan Cedex, France, 2000, pp. 299 – 314.
[3] J. Bisschop, P. Lura, J. van Mier, Shrinkage microcracking in cement-
based materials with low water – cement ratio, Proc. RILEM Int. Conf.
on Early Age Cracking in Cementitious Systems (EAC’01), Haifa,
2001, pp. 79 – 88.
[4] A. Bentur, S. Igarashi, K. Kovler, Control of autogenous shrinkage
stresses and cracking in high strength concretes, Proc. 5th Int. Symp.
on Utilization of High Strength/High Performance Concrete, Sandef-
jord, 1999, pp. 1017 – 1026.
[5] K. van Breugel, J. de Vries, Mixture optimization of HPC in view
of autogenous shrinkage, Proc. 5th Int. Symp. on Utilization
of High Strength/High Performance Concrete, Sandefjord, 1999,
pp. 1041 – 1050.
[6] Ø. Bjontegaard, Thermal dilation and autogenous deformation as driv-
ing forces to self-induced stresses in high performance concrete, Doc-
toral thesis, NTNU Division of Structural Engineering, 1999.
[7] A. Radocea, Autogenous volume change of concrete at very early age,
Mag. Concr. Res. 50 (2) (1998) 107 – 113.
[8] E. Tazawa, Y. Matsuoka, S. Miyazawa, S. Okamoto, Effect of auto-
genous shrinkage on self stress in hardening concrete, Proc. Int.
RILEM Symp. Thermal Cracking in Concrete at early Ages, Munich,
E&FN Spon, London, UK, 1995, pp. 221 – 228.
[9] O.M. Jensen, P.F. Hansen, Influence of temperature on autogenous
deformation and relative humidity change in hardening cement paste,
Cem. Concr. Res. 29 (4) (1999) 567 – 575.
[10] H. Hedlund, J.-E. Jonasson, Effect on stress development of restrained
thermal and moisture deformation, Proc. Shrinkage 2000 —Int. RILEM
Workshop on Shrinkage of Concrete, Paris, RILEM Publications
S.A.R.L. Cachan Cedex, France, 2000, pp. 355 – 375.
[11] J. Bijen, Blast furnace slag cement for durable marine structures, CIP
Royal Library Den Haag, Stichting Betonprisma. ’S-Hertogenbosch,
The Netherlands, 1996.
[12] R.N. Swamy, Holistic design of concrete technology: The only route
to durability and sustainability in construction, Proc. Int. Symp.
Cement and Concrete Technology in the 2000 s, Istanbul, Damla
Matbaacilik Reklamcilik ve Yayincilik Tic. Ltd. S˛ti. Ankara, Turkey,
2000, pp. 58 – 72.
[13] O.M. Jensen, Influence of cement composition on autogenous de-
formation and change of the relative humidity, Proc. Shrinkage
2000 —Int. RILEM Workshop on Shrinkage of Concrete, Paris,
RILEM Publications S.A.R.L. Cachan Cedex, France, 2000,
pp. 143 – 153.
[14] E.A.B. Koenders, Simulation of volume changes in hardening cement-
based materials, Delft University of Technology, PhD, Delft, 1997.
[15] S. Hanehara, H. Hirao, H. Uchikawa, Relationship between autoge-
nous shrinkage and the microstructure and humidity changes at inner
part of hardened cement pastes at early ages, Proc. Autoshrink ’98,
Int. Workshop on Autogenous Shrinkage of Concrete, Hiroshima,
E&FN Spon, London, UK, 1998, pp. 89 – 100.
[16] E. Tazawa, S. Miyazawa, Autogenous shrinkage: What is understood
and which are the further research needs? Proc. Int. Workshop on
Control of Cracking in Early-Age Concrete, Sendai, 2000, (in press).
[17] X. Zhang, Y. Li, K. Wu, Study on autogenous shrinkage and AC
impedance of paste with additives, Proc. Shrinkage 2000 — Int. RI-
LEM Workshop on Shrinkage of Concrete, Paris, RILEM Publica-
tions S.A.R.L. Cachan Cedex, France, 2000, pp. 547 – 557.
[18] K. van Breugel, Simulation of hydration and formation of structure in
hardening cement-based materials, PhD thesis, Technical University
Delft, 1991.
[19] K. Rakel, Beitrag zur Bestimmung der Hydrationswärme von Zement,
PhD, Aachen, 1965.
[20] F. Rostásy, A. Gutsch, M. Laube, Creep and relaxation of concrete at
early ages — experiments and mathematical modelling, Proc. 5th Int.
RILEM Symp. Creep and Shrinkage of Concrete, Barcelona, E&FN
Spon, London, UK, 1993, pp. 453 – 458.
[21] B. Miao, A new method to measure the early-age deformation of
cement based materials, Proc. Shrinkage 2000 — Int. RILEM Work-
shop on Shrinkage of Concrete, Paris, RILEM Publications S.A.R.L.
Cachan Cedex, France, 2000, pp. 381 – 386.
[22] Z.P. Bazant, F.H. Wittmann, Creep and Shrinkage in Concrete Struc-
tures, Wiley, New York, 1982.
[23] Y. Tezuka, J.G. Djanikan, H. Uchikawa, S. Uchida, Hydration char-
acteristics and properties of mixtures of cement and high content of
calcium, Proc. Symp. on Chemistry of Cement, Rio de Janeiro, vol. II,
(1986) 323 – 329.
[24] O.M. Jensen, P.F. Hansen, Water-entrained cement-based materials: I.
Principles and theoretical background, Cem. Concr. Res. 31 (5) (2001)
647 – 654.